♻️ refactor: Improve Ollama response handler

CMakeLists.txt

@@ -56,6 +56,7 @@ add_qtc_plugin(QodeAssist
     providers/OllamaProvider.hpp providers/OllamaProvider.cpp
     providers/LMStudioProvider.hpp providers/LMStudioProvider.cpp
     providers/OpenAICompatProvider.hpp providers/OpenAICompatProvider.cpp
+    providers/OllamaMessage.hpp providers/OllamaMessage.cpp
     QodeAssist.qrc
     LSPCompletion.hpp
     LLMSuggestion.hpp LLMSuggestion.cpp

providers/OllamaMessage.cpp

@@ -0,0 +1,76 @@
+/* 
+ * Copyright (C) 2024 Petr Mironychev
+ *
+ * This file is part of QodeAssist.
+ *
+ * QodeAssist is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * QodeAssist is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with QodeAssist. If not, see <https://www.gnu.org/licenses/>.
+ */
+
+#include "OllamaMessage.hpp"
+
+namespace QodeAssist::Providers {
+
+OllamaMessage OllamaMessage::fromJson(const QJsonObject &obj, Type type)
+{
+    OllamaMessage msg;
+    msg.model = obj["model"].toString();
+    msg.createdAt = QDateTime::fromString(obj["created_at"].toString(), Qt::ISODate);
+    msg.done = obj["done"].toBool();
+    msg.doneReason = obj["done_reason"].toString();
+    msg.error = obj["error"].toString();
+
+    if (type == Type::Generate) {
+        auto &genResponse = msg.response.emplace<GenerateResponse>();
+        genResponse.response = obj["response"].toString();
+        if (msg.done && obj.contains("context")) {
+            const auto array = obj["context"].toArray();
+            genResponse.context.reserve(array.size());
+            for (const auto &val : array) {
+                genResponse.context.append(val.toInt());
+            }
+        }
+    } else {
+        auto &chatResponse = msg.response.emplace<ChatResponse>();
+        const auto msgObj = obj["message"].toObject();
+        chatResponse.role = msgObj["role"].toString();
+        chatResponse.content = msgObj["content"].toString();
+    }
+
+    if (msg.done) {
+        msg.metrics
+            = {obj["total_duration"].toVariant().toLongLong(),
+               obj["load_duration"].toVariant().toLongLong(),
+               obj["prompt_eval_count"].toVariant().toLongLong(),
+               obj["prompt_eval_duration"].toVariant().toLongLong(),
+               obj["eval_count"].toVariant().toLongLong(),
+               obj["eval_duration"].toVariant().toLongLong()};
+    }
+
+    return msg;
+}
+
+QString OllamaMessage::getContent() const
+{
+    if (std::holds_alternative<GenerateResponse>(response)) {
+        return std::get<GenerateResponse>(response).response;
+    }
+    return std::get<ChatResponse>(response).content;
+}
+
+bool OllamaMessage::hasError() const
+{
+    return !error.isEmpty();
+}
+
+} // namespace QodeAssist::Providers

providers/OllamaMessage.hpp

@@ -0,0 +1,71 @@
+/* 
+ * Copyright (C) 2024 Petr Mironychev
+ *
+ * This file is part of QodeAssist.
+ *
+ * QodeAssist is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * QodeAssist is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with QodeAssist. If not, see <https://www.gnu.org/licenses/>.
+ */
+
+#pragma once
+
+#include <QDateTime>
+#include <QJsonArray>
+#include <QJsonObject>
+#include <QObject>
+
+namespace QodeAssist::Providers {
+
+class OllamaMessage
+{
+public:
+    enum class Type { Generate, Chat };
+
+    struct Metrics
+    {
+        qint64 totalDuration{0};
+        qint64 loadDuration{0};
+        qint64 promptEvalCount{0};
+        qint64 promptEvalDuration{0};
+        qint64 evalCount{0};
+        qint64 evalDuration{0};
+    };
+
+    struct GenerateResponse
+    {
+        QString response;
+        QVector<int> context;
+    };
+
+    struct ChatResponse
+    {
+        QString role;
+        QString content;
+    };
+
+    QString model;
+    QDateTime createdAt;
+    std::variant<GenerateResponse, ChatResponse> response;
+    bool done{false};
+    QString doneReason;
+    Metrics metrics;
+    QString error;
+
+    static OllamaMessage fromJson(const QJsonObject &obj, Type type);
+
+    QString getContent() const;
+
+    bool hasError() const;
+};
+
+} // namespace QodeAssist::Providers

providers/OllamaProvider.cpp

@@ -25,6 +25,7 @@
 #include <QNetworkReply>
 #include <QtCore/qeventloop.h>
 
+#include "OllamaMessage.hpp"
 #include "logger/Logger.hpp"
 #include "settings/ChatAssistantSettings.hpp"
 #include "settings/CodeCompletionSettings.hpp"
@@ -87,53 +88,41 @@ void OllamaProvider::prepareRequest(QJsonObject &request, LLMCore::RequestType t
 
 bool OllamaProvider::handleResponse(QNetworkReply *reply, QString &accumulatedResponse)
 {
-    QString endpoint = reply->url().path();
-
-    bool isComplete = false;
-    while (reply->canReadLine()) {
-        QByteArray line = reply->readLine().trimmed();
-        if (line.isEmpty()) {
-            continue;
-        }
+    const QString endpoint = reply->url().path();
+    auto messageType = endpoint == completionEndpoint() ? OllamaMessage::Type::Generate
+                                                        : OllamaMessage::Type::Chat;
+
+    auto processMessage =
+        [&accumulatedResponse](const QJsonDocument &doc, OllamaMessage::Type messageType) {
+            if (doc.isNull()) {
+                LOG_MESSAGE("Invalid JSON response from Ollama");
+                return false;
+            }
 
-        QJsonDocument doc = QJsonDocument::fromJson(line);
-        if (doc.isNull()) {
-            LOG_MESSAGE("Invalid JSON response from Ollama: " + QString::fromUtf8(line));
-            continue;
-        }
+            auto message = OllamaMessage::fromJson(doc.object(), messageType);
+            if (message.hasError()) {
+                LOG_MESSAGE("Error in Ollama response: " + message.error);
+                return false;
+            }
 
-        QJsonObject responseObj = doc.object();
+            accumulatedResponse += message.getContent();
+            return message.done;
+        };
 
-        if (responseObj.contains("error")) {
-            QString errorMessage = responseObj["error"].toString();
-            LOG_MESSAGE("Error in Ollama response: " + errorMessage);
-            return false;
-        }
+    if (reply->canReadLine()) {
+        while (reply->canReadLine()) {
+            QByteArray line = reply->readLine().trimmed();
+            if (line.isEmpty())
+                continue;
 
-        if (endpoint == completionEndpoint()) {
-            if (responseObj.contains("response")) {
-                QString completion = responseObj["response"].toString();
-                accumulatedResponse += completion;
+            if (processMessage(QJsonDocument::fromJson(line), messageType)) {
+                return true;
             }
-        } else if (endpoint == chatEndpoint()) {
-            if (responseObj.contains("message")) {
-                QJsonObject message = responseObj["message"].toObject();
-                if (message.contains("content")) {
-                    QString content = message["content"].toString();
-                    accumulatedResponse += content;
-                }
-            }
-        } else {
-            LOG_MESSAGE("Unknown endpoint: " + endpoint);
-        }
-
-        if (responseObj.contains("done") && responseObj["done"].toBool()) {
-            isComplete = true;
-            break;
         }
+        return false;
+    } else {
+        return processMessage(QJsonDocument::fromJson(reply->readAll()), messageType);
     }
-
-    return isComplete;
 }
 
 QList<QString> OllamaProvider::getInstalledModels(const QString &url)